Battery Fuel Gauge

One of the main tasks of the smart battery is to establish communication between the battery and user. A fuel gauge that indicates state-of-charge is part of this effort. When pressing the TEST button on a fully charged SMBus battery, all signal lights illuminate. On a partially discharged battery, half the lights illuminate, and on an empty battery all lights remain dark or a red light appears. Figure 1 shows a fuel gauge of a battery that is 75 percent charged, with three lights glowing.

State-of-charge readout of a “smart” battery

Figure 1: State-of-charge readout of a “smart” battery

Signal lights indicate the battery SoC when pressing the TEST button.

Courtesy of Cadex

While the SoC information displayed on a battery or computer screen is helpful to the user, it does not assure sufficient runtime, because the fuel gauge resets to 100 percent on a full recharge regardless of how much capacity the battery can store. A serious miscount occurs if an aged battery shows 100 percent SoC while the battery’s ability to hold charge has dropped to 50 percent or less. We ask, “100 percent of what?” If, for example, 100 percent of a good battery results in a four-hour runtime, the same indication with a battery holding half the capacity would run for only two hours. The user should know that the fuel gauge only shows SoC and capacity is missing.

Other than using coulomb counting and measuring internal resistance, there is no reliable method to calculate the state-of-health (SoH) of a battery; current technologies provide only rough estimations. However, there is a way to display the capacity of a smart battery digitally.

At time of manufacture, the SMBus battery is programmed with a specified capacity, which is 100 percent by default, and the battery keeps this information as permanent data. With each full charge, the battery resets to the charge flag; and during discharge the coulomb counter measures the consumed energy. A perfect battery would deliver 100 percent on a calibrated fuel gauge. As the battery ages and the capacity drops, the delivered energy between charges decreases. The discrepancy between the factory-set 100 percent and the delivered coulombs after a full charge can thus determine the battery capacity.

The SoC and capacity information can be shown on a linear display using colored LEDs. The green lights indicate the usable capacity; the empty part of the battery is marked with dark LEDs; and the unusable part is shown with red LEDs. Figure 2 illustrates a tri-state fuel gauge. Alternatively, the results can be a shown on a digital display. 

Tri-state fuel gauge

Figure 2: Tri-state fuel gauge.The tri-state fuel gauge reads the “learned” battery information on the SMBus and displays it on a multi colored LED bar. The illustration shows a partially discharged battery of 50% SoC with 20% empty and 30% unusable.

Courtesy of Cadex

Installed in an SMBus charger, the tri-state fuel gauge indicates the usable capacity of SMBus compatible batteries and gives the user a clear indication of when to retire a pack. The pass/fail level is user-selectable — at, say, 60, 70 or 80 percent — depending on the minimum required performance level. A failed battery prompts the user to first calibrate the battery. If the charge/discharge/charge cycle did not raise the capacity to the set target level, a red light on the charger indicates a fail. A lit green ready light always assures that the battery has been successfully charged and meets the capacity requirements. (Cadex offered such a charger but the buyers did not appreciate the benefit.)

Cars with electric propulsion systems use similar fuel-gauge technologies to determine the state-of-charge and state-of-function of the battery. The challenge is in knowing how far a vehicle can travel with a fully charged battery in various ambient conditions. A tailwind on a sunny day is more forgiving than a headwind with snow. The aging of a battery adds to the challenge and the fuel gauge will lose accuracy over time. Showing 30 minutes of remaining runtime and then shutting down, as is common with laptops, will not be easily forgiven with the electric vehicle.

Comments

On December 22, 2010 at 10:15pm
BWMichael wrote:

Battery memory sucks!

On January 1, 2011 at 8:14pm
Steve Arey wrote:

Thanks for that information.  Do you answer questions?

On January 3, 2011 at 8:50pm
BWMichael wrote:

I am not an official battery university ‘question answerer’ but if u have a question put it on here and i will answer if i can

On January 10, 2011 at 9:26pm
Ganeshkumar GK wrote:

Hi, Am an Administration Manager in a 18 hole golf course in India. I have a fleet of 88 golf carts.  Presently am in the process of replacing the batteries in the carts as they have started to give away.

Can you suggest some ways and means for maximising the battery life ? Am doing a lot of trial and error now to arrive at the battery configuration as the OEM batteries are not in production now.

Would be glad to hear from you.

Thanks and Regards

On March 9, 2011 at 10:01pm
Wal Sakaluk wrote:

Hello Ganesh, It is hard to specifically advise you without knowing the environmental and operating conditions of your battery operation. Golf carts, typically, deploy lead acid batteries, these can be conventional “wet” cells were the electrolyte level and condition can be monitored or sealed gas recombination technology batteries. I will give you life prolonging recommendations that are general for this type of battery.
1. Do not deep discharge the batteries
2. Do not subject the batteries to mechanical shock
3.Charge the batteries correctly
4.Cooler is better
5.All connections MUST be clean and secure and kept that way

The OEM battery would have been of an industry standard form factor and replication of function and specification could be easily made with locally procurable batteries IN CONSULTATION WITH A QUALIFIED AND EXPERIENCED locally sourced electrical engineer. Do not take the advice of a battery vendor (unless they have an engineering capacity) or self taught mechanic. Trial and error is a worry, seems like you don’t have the right person helping you. This is basic electrical engineering but a precise science none the less. Get professional help to save this wasteful practice and to ensure you don’t damage your assets or injure anybody. Easy if you know what you are doing. Play safe!!

On July 6, 2011 at 2:03pm
Joe Accetta wrote:

We have developed an true state of charge sensor for lead acid batteries with addtional features such as cell temperature etc. Preproduction prototypes are available. The sensor is packaged in a conventional battery cap and is a direct replacement for existing caps with standard bayonet base. If you have a comment or application for this development please contact me.

On July 23, 2011 at 9:38pm
Steve Arey wrote:

Is “we” mean the BAttery University?

On July 24, 2011 at 6:15am
Joe Accetta wrote:

Steve,
This is a development of JSA Photonics. www.jsaphotonics.com. I am not sure how the battery fuel gauge described on the Battery University site works. Our sensor determines the true specific gravitiy of the electrolyte.

On October 30, 2011 at 9:25pm
Vincent Lopez wrote:

I have an automotive battery, sealed type with a built in hydrometer that I charge using a small automatic charger. One time the battery went flat and when i attempted to charge it, itwon"t take a charge. So I brought it to a charging station, the owner of the station said he will use a rapid charger to quickly charge the battery for one hour. After an hour, the owner said its already charge but not yet full, but I took the battery anyway and continued to charge it at home using my trusty charger overnight. Morning came and the battery was already hot and there was a hissing sound, turn off the charger and connected the battery in my car but still it won’t start. Should I discard the battery or continue charging it?

On October 30, 2011 at 10:07pm
Wal Sakaluk wrote:

Vincent,
Your battery is history. Sell it for scrap.

On October 31, 2011 at 7:17am
Joe Accetta wrote:

Wal is correct. Your battery is lprobably gone.

On December 25, 2011 at 4:45pm
Francesco wrote:

Does the SMBus work also on electric cars? If no, why? We are developing an car sharing with electic cars and we wonder how the central system can retrieve informations about SOC and SOH of the several and different cars within the car sharing system..
Thanks for the help

On March 9, 2012 at 10:05am
Bob White wrote:

I need a fuel gauge for a seven-cell series LiPo battery (12-cell growth ability would be a groove).  Normal discharge current is 65 Amps.  I have been unable to find such a device.  Battery monitors are available, but fuel gauges appear to be limited to about four cells.  Am I searching for the wrong devices?  I would think that the fuel gauge might not have to care how many cells are in the battery.

Thanks,
Bob

On March 9, 2012 at 10:39am
Joe Accetta wrote:

Bob,
What you are looking might be described as the holy grail for the automotive fuel guage of the future. We have not tried our technology in any solid state battery configuration. It has only been tested on open port( liquid electrolyte) lead acid cells and would probably work in any liquid cell. Sorry we cant help at this time..

On March 12, 2012 at 4:33am
Karthick wrote:

Hi..,

I have One doubt any one give This
Case 1:when i use 100A 12V Battery at a Lode 15A, Again i connect one dynamo with battery charger,
what is the time taken for charging.
Case 2: 100A battery is discharged 80% how match time taken for full charge

On June 15, 2012 at 4:56am
Wang YJ wrote:

Hi Bob:
Please check this battery guage, BG2-XX, it can suit for 12s battery pack
http://www.szgrn.com/battery-gauge.html
Hope it is useful.

On July 24, 2012 at 2:17am
Kevin Driscoll wrote:

Hi,
I have just replaced 10 batteries (four Starter and six domestic) on my boat together with a new battery charger.  My old batteries were getting hot and gassing.  I would like to monitor the new batteries.  Any suggestions? I leave my boat for about five months a year, is it wise to leave the battery charger on all the time?
Grateful for any help.
Regards
Kevin

On July 24, 2012 at 2:38am
Wal Sakaluk wrote:

Lead acid batteries do not like to just sit around.

Float charging will prolong their life.

On August 5, 2012 at 1:56am
Kevin Driscoll wrote:

Thanks Wal

On August 9, 2012 at 9:44am
yiwen wrote:

I need to write some software to monitor battery state of health. The battery pack I have is 14 serial-connected lithium-ion batteries. I did some test and get the ampHour vs voltage plot for the first full-discharge interation (start from 4.1V@26ampH and end at 2.8V@0ampH). What I am thinking is to monitor battery voltage at run time and use coulomb counting to get ampH. Compare the actual ampH data with the ideal ampH data in the plot.

So the SOH should be actual ampH divided by 1st interation ampH. Does it sound to be a reasonable way to estimate SOH?

On January 3, 2013 at 5:16am
vijay dheeraj wrote:

Needless to introduce cairn india ltd is oil & gas E&P company in INDIA.
We have 8 offshore platforms and all the instruments are opearted at 24V DC which are powered by 24V battery bank with Solar panel as a source of energy.
We have been using sealed lead acid batteries of 12V*120AH for the past 10 years but we have been facing the battery failure once in 2 years.
We have also used GEL batteries of sonnenschein make of 2v*960AH and have been using 2v*1130AH batteries since a month as they have drained.
Please suggest which type of battery is best for solar & wind applications on offshore oil & gas platforms i.e., on hazardous area applications.
I have also learned that AGM is more efficient and also has long life compared to SLA and GEL.
Please suggest which is the best battery fo this application.
I have gone through variour articles on net and found GEL & AGM batteries are efficient enough for solar and wind applications.

On February 28, 2013 at 6:21pm
jeff gerhard wrote:

vijay dheeraj- Do you use a proper charge controller? How are you cycling the batteries?
These batteries should not be “deep cycled”.  Study the literature for lead acid batteries.

On November 11, 2013 at 9:51pm
Alfonso wrote:

I wonder if you can help me on this issue.
In a bank of 12 2v batteries which are connected in serial to give 24v.
How fast does a battery that doesn’t work correctly beguin to affect the other batteries in the bank? Is it a matter of hours, days or months before the other batteries start working incorrectly too?

thank you for your help.

On November 11, 2013 at 11:09pm
Wal Sakaluk wrote:

It depends on the modality of the fault in that single cell.

If it fails short circuit then the other cells will now be subjected to a higher charging voltage. The other cells will soon fail.

If it fails open circuit then the whole bank is rendered useless, however no harm, other than deterioration due to being idle will befall the others.

If the internal resistance of that one cell is too high then the other cells will have a lower charging voltage and full autonomy will never be achieved.

Periodic individual cell voltage monitoring is important to ensure early detection of these fault or deterioration modalities and to thentake the appropriate action.